Fuses: Why aren't they obsolete?

[SeanB]

So thankful I work where you only have a single supply system, triplicated for 3 phase, but still the same thing. Not the US mix of however many they have, single phase, split phase, assorted 3 phase varieties, high leg delta and all with similar looking voltages and scarily enough various floating voltages on the conductors.

Here 230VAC to neutral all round, 400VAC between phases and current is going to be the only differentiator for the breakers. The incoming ones are a lot bigger, better rated and have to live being fed by a great big 200kVA transformer via some 200A fuses, but downstream there is a lower requirement from the reduced cable size providing increased source impedance. Biggest breaker is the lift, 100A and cables to match, but the draw is 20A or so per phase, but the supply has to be capable of providing the all important overrun test motor current without tripping before the motor room dashpot breaker.

You can have whatever mechanical, electronic or semiconductor overcurrent devices you want, but always upstream of them there will be a fuse of some kind, there as the in case all else fails device, and generally if it goes for anything other than old age something downstream is going to be really badly borked.

[krish2487]

This!
Fuses have been around for so long. They are well characterized and have specified response times and capacities. Fast blow, slow blow, delayed trigger, HRC so on. Each has a specific application and characteristics. A single semiconductor alternative will not suit all the use cases.


Second, variance in semiconductors. Granted, todays manufacturing has the tolerance to a predictable limit. Then factor in the accumulated tolerances of other semiconductors and passives. Now the net tolerance of the system in terms of response time, capacity, MTBF might or might not be with in the required spec of the application.


All these factors make a semiconductor replacement unviable for a repeatable, predictable behavior. Atleast for the technology available today.


Quote from: oldway on Yesterday at 09:19:16 AM

There is a lot of reason why the fuse is a must-have component
1) Its cost is very low
2) it is a very simple device and therefore not susceptible to breakdowns
3) it is able to interrupt very high currents in a small volum

e

[Wytnucls]

Can't you build a fast enough current sensor and semiconductor switch that could 100% of the time protect both the switch and the attached equipment from overcurrent conditions?  Is tit not practical to do this or is it just too expensive?  Worst case, you could always also have a fuse in-line, but it would never blow on you.  Wouldn't this be ideal for DMMs, but also a huge range of other equipment?  Thanks.

If I remember correctly, IEC61010 does not allow semi-conductors being the only protection againts fire / explosion.

The Gossen MetraHit 30M, as a precision instrument, only measures low currents (100mA max).
The original design had a soldered surface mount fuse, which blew so often due to careless handling, that the manufacturer eventually replaced it with a resettable polyswitch fuse on a later model revision.

I am surprised they could fit a polyswitch on a CAT rated meter. I can't say I have ever seen a polyswitch that would be good for CAT rated applications above 300V.

I'm not sure which PPTC they used, but some of them are rated for 600V/1A fault voltage and current. The impedance of the mA shunt circuit is 14 Ohms on the 100mA range. There is no Amp range.

[james_s]

How about a polyswitch in series with a slightly higher rated fuse? A big overload will blow the fuse, but the polyfuse will take care of the usual careless mistakes.

[Wytnucls]

That's a possibility, but is not mentioned by Gossen. I'll see if I can trace the protection components when I have some time.

[james_s]

Oh it doesn't really matter to me how it's done in that particular instrument, that's just how I would be inclined to design one.

[Vtile]

It is the most robust solution.

Install and forget.

PS. I forgot, fuses aren't for direct equipment or human safety, their main purpose is to prevent fire and further collateral damage from the failed system. This in electrical perspective, then low energy electronics might look at them more of an equipment perspective.